002&7519/92 $5.00 + 0.00 Pergumon Press Ltd Q 1992 Australian Society for Parasitology

Interwtional Journaifor Parasitology Vol. 22, No. 4,pp. 527-531, 1992 Printed in Great Brimin

DETECTION OF ~PI~~~~~C~~S ~~~~~~~~~ ANTIGENS IN STOOLS USING SPECIFIC MONOCLONAL ANTIBODY W. CHAICUMPA,*L. YBANEZ,* V. KITIKOON,~S. PUNGPAK,~Y. RUANGKUNAPORN,* M. CHONGSA-NGUAN*and S. SORNMANI~ *Department of Microbiology and Immunology, TDepartment of Tropical Medicine and SDepartment of Clinical Tropical Medicine, Faculty of Tropical Medicine, Mahidol University, 420/6 Rajvithi Road, Bangkok 10400, Thailand (Received 29 January 1992;accepted 9 March 1992) Abstra&--&A~cuiuPA W., YBANEZL., KIIXOON V., P~JNGPAK S., RUANGKUN~RN Y., CHONGSA-NGUAN M. and SORNMANIS. 1992.Detection of Opisthurchis viverrini antigens in stools using specific monoclonal antibody. International Journalfor Parasitology 22: 527-531. A sandwich enzyme-linked immunosorbent assay (ELISA) was developed for detecting Opisthorcbis viverrini antigen in faecal extracts of four groups of individuals. These were 24 patients with 0. viverrini infection only (group l), 31 patients with 0.

viverrini and other parasitic infections (group 2), 141 patients with other parastitic infections (group 3) and 21 normal, parasite-free individuals (group 4). The first antibody used in the ELISA was polyclonal immunoglobulin G prepared from the serum of a rabbit previously immunized with crude extract of 0. viverrini. The second antibody was monoclonal antibody specific to an antigen located in the worm tegument and muscular tissue. Sensitivity of the assay was 31% while specificity was 100%. Considerations for improving the sensitivity are discussed. INDEX KEY WORDS: I~un~ia~osis: intestinal flukes: rabbit anti-O. viverrini serum.

OF~st~arehis viverrini; ELISA: opisthorchiasis;

THE

small

faecal specimens of patients, using specific monoclonal antibody, are reported herein. Faecal specimens were collected from four groups of adult individuals: 24 patients with 0. viverrini only (group l), 31 patients with 0. viverrini and other parasitic infection(s) (group 2) (Table l), 141 patients with other parasitic infection(s) (group 3) (Table 2) and 21 normal, parasite-free individuals (group 4). Subjects of groups 1, 2 and 3 were patients admitted to the Bangkok Hospital for Tropical Diseases. Those belonging to groups 3 and 4 were from areas non-endemic for opisthorchiasis. Stools were examined using Stoll’s egg count technique (Stoll, 1923) and formalin-ether concentration. Faecal samples were collected and examined individually on 3 consecutive days and two smears were examined from each faecal specimen. Patients who were allocated to group 1 were those whose stool specimen revealed 0. viverrini eggs. In order to confirm that these patients were infected only with 0. viverrini they were subsequently treated with praziquantel (40 mg kg-’ body weight) and saturated magnesium sulphate (45 ml) orally 7 h after the treatment with praziquantel. Stools were collected within 12 h

clinical manifestations of liver fluke infection caused by Opisthorchis viverrini are not pathognomonic, so diagnosis is based mainly on laboratory examination, to demonstrate eggs in stools, duodenal fluid or bile (Harinasuta, 1969; Felmeier, Feldheim,

Rasp & Bienzle, 1981). However, in the case of light infection, examination for eggs in stools is uncertain. Also 0. viverrini eggs are difficult to distinguish from eggs of small intestinal flukes like Pro~thodendrium ~h~~ropso~us bonnei, ~aplorchis mo~en~ampi, t~i~h~i and H. p~rni~io (Radomyos, Bunnag & Harinasuta, 1984). In heavy infection where there is

biliary obstruction, the eggs remain in the bile duct and gall bladder and cannot be recovered in the faeces (Evans, Bourgeois, Comer & Keschamaras, 1971; Flavell, Patanapanyasat, Lucas & Vongsangnak, 1980; Harinasuta, Riganti & Bunnag, 1984). In these instances, cannulated bile is the specimen of choice for demonstration of eggs. However, cannulation of the biliary tract is not a simple procedure and may pose some danger. Immunological assays are altematives for these sit~tions. Experimental data on the development of an en~me-Iinked immunosorbent assay (ELISA) to detect 0. viverrini antigens in 527

W. CHAICUMPA et al.

528 TABLE

I-LISTOF

PARAS~TESOTHERTHAN PATIENTSOFGROUP

0. viverrini FOUNDIN 2

Number of patients

Name(s) of parasite(s)

1

Blastocystis hominis Echinostoma spp. Hookworm(s) Hookworm(s) and B. hominis Hookworm(s) and Echinostoma spp. Hookworm(s) and Endolimax nana Hookworm(s) and Entamoeba coli Hookworm(s) and Giardia lamblia Hookworm(s) and Strongyloides slercoralis Hookworm(s), B. hominis and Echinostoma spp. Hookworm(s), B. hominis and E. nana Hookworm(s), B. hominis and S. stercoralis Hookworm(s), B. hominis, E. coli and S. stercoralis Hookworm(s), Echinostoma spp. and G. lamblia S. stercoralis Taenia saginata Trichomonas hominis

I 11 1 1 2 1 1 2 1

1 2 1 1 2 1 1

TABLET-LISTOFPARAS~TESFOUNDINPATIENT~OFGROUP

Number of patients

Name(s) of parasites

1*

Ascaris lumbricoides, hookworm(s), Plasmodium falciparum and S. stercoralis A. lumbricoides, hookworm(s), P. falciparum and Trichuris trichiura A. lumbricoides. P. falciparum, T. hominis and T. trichiura A. lumbricoides and T. trichiura B. hominis, hookworm(s) and P. falciparum Entamoeba histolytica Filaria Gnathostoma spinigerum Hookworm(s) Hookworm(s), G. lamblia and P. falciparum Hookworm(s) and P. falciparum Hookworm(s), P. falciparum and S. siercoralis Hookworm(s) and S. srercoralis Hookworm(s), S. stercoralis and T. rrichiura Paragonimus heterotremus P. falciparum P. falciparum and S. stercoralis P. falciparum. S. siercoralis and T. saginata P. falciparum. S. siercoralis, T. hominis and yeast Schistosoma japonieum S. stercoralis T. saginata Trichinella spiralis T. trichiura

2

I 1 I 3 1 3 12

1 4 2

1 1 21 17

I I 1 10 21 13 21 1

(s)

* Stool extract of this patient antigen by the ELBA.

thereafter. Sedimentation (Faust, Sawitz, Tobie, Odom, Peres & Lincome, 1939) was carried out on the stool specimens for recovering parasite(s). Patients with only 0. viverrini worms found in their stools were allocated to group 1. Faecal samples of all groups were extracted individually using 0.1 M-titrate buffer pH 5.0 containing 5% bovine serum albumin (BSA). To 1 g of specimen, 4 ml of buffer was added, mixed thoroughly, then centrifuged at 200 x g for 10 min. The supernate was collected and stored in 1 ml until used in the ELISA for aliquots at -20°C detecting 0. viverrini antigen. Previous experiments indicated that the soluble 0. viverrini antigen was not precipitated even when the buffered stool sample was centrifuged at 10,000 x g. Laboratory-bred freshwater fish (Pontius spp.) were infected with cercariae of 0. viverrini shed from infected Bithynia snails. laboratory-bred, Metacerariae were dissected out from the muscles of the infected fish 6 weeks after infection and golden

3

was positive for 0. viverrini

were fed 100 metacercariae each. Two months after infection, the hamsters were sacrificed by ether euthanasia. The livers were dissected out, placed between two glass plates and examined for adult 0. viverrini. The parasites were removed from the bile ducts, washed with sterile normal saline solution (NSS) and finally with distilled water (DW). The worms were ground in DW containing protease inhibitors, namely 0.1 mM-phenylmethylsulphonylfluoride (PMSF), 0.1 mM-tosyl-amide-2-phenylethylchloromethyl ketone (TPCK) (Sigma Chemical Co., U.S.A.) and 10 mM-ethylenediamine tetraacetic acid (EDTA) in a glass tissue grinder. The preparation was ultrasonicated five times at 20 kHz at 4°C for 10 mitt, followed by centrifugation at 10,000 x g at 4°C for 10 min. The supernatant was collected and the protein content was determined (Lowry, Rosebrough, Farr & Randall, 1951). The supernatant (0. viverrini antigens) was aliquoted and stored at - 20°C until used. hamsters

Research Note Rabbit anti-O. viverrini immunoglobulin G was prepared by immunizing a 3 kg rabbit intramuscularly with 1 mg of the 0. viverrini antigens in 0.6 ml of NSS mixed with an equal volume of complete Freund’s adjuvant (Difco Laboratories, U.S.A.). Immunization was repeated two more times at 14-day intervals but incomplete Freund’s adjuvant (Difco Laboratories, USA) was used in the booster doses. Two weeks after the last injection, the animal was bled and the antibody titre of the serum was assessed against homologous antigens by indirect ELISA (Indrawati, Chaicumpa, Setasuban & 1991). Total immunoglobulins Ruangkunaporn, were extracted from the serum by precipitation three times with 50% saturated ammonium sulphate. The pellet obtained from the last precipitation was dissolved in a small amount of DW, dialysed against DW and the protein content was determined. Immunoglobulin G (IgG) was separated out from the total immunogobulins by affinity chromatography using protein A-Sepharose CL4B (Pharmacia Fine Chemicals, Sweden). Washing and eluting buffers were 0.1 M-phosphate buffer pH 8.0 and 0.1 Mglycine-HCl pH 2.8, respectively. The IgG eluted from the column was dialysed against DW and centrifuged at 10,000 x g at 4°C for 10 min. The protein content of the supernatant was determined and the total IgG preparation was stored at -20°C in aliquots. A monoclonal antibody, 0V53 B,, which was specific to the antigen located in the worm tegument and muscles (Chaicumpa, Ruangkunaporn, KalamLimavongpranee, Kitikoon, Khusmith, baheti, Pungpak, Chongsa-nguan & Sornmani, 1991) was utilized as second antibody in the sandwich ELISA for detecting 0. viverrini antigen in the faecal specimens. Optimal dilutions and incubation parameters of solid-phase antibodies and the peroxidaseconjugated rabbit anti-mouse immunoglobulins (Dakapatt, Denmark) for the sandwich ELISA were determined by checkerboard titrations. Wells of microtitre ELISA plates (Greiner Company, Germany) were coated with 20 pg ml-’ of rabbit anti-O. viverrini total IgG in carbonate-bicarbonate buffer pH 9.6 and the plates were incubated at 37°C until dry. One hundred microlitre volumes were used throughout the assays. The wells were blocked with a 1% (w/v) suspension of BSA in phosphate buffered saline (PBS), pH 7.4 at 37°C for 1 h. The wells were aspirated and faecal supernatants were added to duplicate precoated wells. After incubation for 1 h at 37°C the wells were washed with 0.05% (v/v) Tween20/PBS and lightly tamped dry. To each well was added a fixed amount (64 indirect ELISA units:

529

Chaicumpa, Thin-inta, Khusmith, Tapchaisri, Echeverria, Kalambaheti & Chongsa-nguan, 1988) of the specific monoclonal antibody, incubated for 1 h at 37”C, washed as above and all wells were coated with rabbit anti-mouse immunoglobulin-peroxidase conjugate. After incubation and washing as above, the chromogen substrate was added and allowed to develop for 30 min, and the reaction was stopped with 1 N-NaGH. The absorbance at 495 nm was measured on a Uniscan II microELISA plate reader (Labsystems, Finland). Sensitivity and specificity of the assay were calculated using the method of Galen (1980). It was found that the mean absorbance (8 of the faecal extracts of the normal individuals (group 4) was negligible (0.003), with a standard deviation (SD) of 0.004. When the cutoff absorbance was made at 8 + 3 SD., which was 0.015, eight of the 24 specimens (33%) from group 1, and nine of the 31 specimens (29%) from group 2, were positive for 0. viverrini antigens. One specimen from group 3 and none from group 4 was positive at this cutoff limit. Thus the sensitivity and specificity of the assay were 31 and 98%, respectively. The sandwich ELISA was used also to determine the lowest concentration of the 0. viverrini antigens that could be detected. Antigens of known concentrations in faecal supernatant of normal, parasite-free individuals were used. The smallest amount of antigen which could be detected by the ELISA was 250 ng. Sensitivity of any sandwich enzyme assay system depends on the immunological reactivity of the antibodies used (Chaicumpa et al., 1988). In our ELISA, the rabbit anti-O. viverrini IgG used for sensitization of the solid phase was potent since the serum from which it was extracted was hyperimmune. The second antibody was monoclonal which would have reacted with only one epitope of the antigen which might be present only in a small fraction of the minute amount of antigen in a patient’s stool extract. Between the two antibodies, it is likely that the monoclonal antibody accounted for the low sensitivity. An additional consideration which might influence the sensitivity of antigen detection is the presence of interfering materials in the faecal samples (Hovi, Vaisanen, Ukkonen & von Bonsdorff, 1982; Hanvanich, Viscidi, Laughon, Bartlett & Yolken, 1985; Green, 1986; Kinsley, Prashad & Pickering, 1986). These materials cause denaturation, proteolysis or desorption of reactants on the solid phase surfaces leading to a markedly decreased reactivity in the ELISA. Most of the investigators mentioned above found the interfering substance to be a protease. In our study, 0.1 M-citrate buffer pH 5.0 containing 5% BSA was used in the

530

W. CHAICUMPAet al.

preparation of stool samples. The acid buffer was intended to decrease the activity of pH-specific intestinal proteases and BSA was added to provide an excess of substrate. This was one of the two diluents recommended by Viscidi, Laughon, Hanvanich, Bartlett & Yolken (1984) and Yolken (1985) to reduce desorption. Fifty per cent foetal calf serum is more effective but was not chosen because of its expense. Sensitivity of the antigen detection system might be improved by using a different solid phase such as nitrocellulose in a dot-blot ELISA. Nitrocellulose has a high binding capacity and permits the detection of less abundant antigens in the sample. It binds approximately 1000 times more proteins per surface area than the microplate (Harlow & Lane, 1988). It would alleviate the problem of desorption, since the faecal specimen can be directly dotted on the nitrocellulose, obviating the need for coating with polyclonal antibodies as in our ELISA. However, our attempt to detect 0. viverrini antigens using a dot-blot ELISA was not successful, as faecal extract of normal individuals produced high backgrounds indistinguishable from true positive reactions (unpublished observation). The only specimen in group 3 (mixed infection of hookworms, A. lumbricoides, P. falciparum and S. stercoralis) which produced a positive reaction in the ELISA, was from an individual who, while presently residing in a non-endemic area (Kanchanaburi Province, about 200 km west of Bangkok) had recently migrated from the northeast of Thailand, where opisthorchiasis is endemic. The faecal extract of this patient may have contained 0. viverrini antigens, even though 0. viverrini eggs could not be found. The result of the antigen detection of this case should not be regarded as a false positive as other individuals of the same group who were infected with the same species of heterologous parasites were negative for 0. viverrini antigen. The ELISA developed in our study had low sensitivity compared to the microscopic examination for eggs. This is not altogether unexpected as the monoclonal antibody used in the system was directed to worm somatic antigens which are perhaps excreted in the faeces in insufficient quantities. It is possible that monoclonal antibodies to the parasite’s metabolic or egg antigens (which would be more readily available in stools) could be more sensitive. Attempts along this line have recently been conducted by Sirisinha and his colleagues (Sirisinha, Chawengkirtikul, Sermswan, Amompant, Monkolsuk & Panyim, 1991) using a monoclonal antibody against a metabolic antigen of the parasite. The sensitivity of their test (57%) was slightly better than that reported herein. We are currently producing a

monoclonal antibody to a parasite egg antigen for a similar purpose. It could be speculated that the antigen detection assay using a combination of specific monoclonal antibodies (those directed against somatic antigens, metabolic antigens and egg antigens) would have higher sensitivity than a test using only one monoclonal antibody. Acknowledgements-This work received financial support from the International Development Research Centre (IDRC), Canada. The authors thank Dr May Ho, University of Calgary, Canada for her helpful discussion and review of the manuscript. Thanks are extended to Miss Jatuporn Chaiyawan for typing the manuscript. REFERENCES CHAICUMPAW., THIN-INTAW., KHUSM~THS., TAPCHAISRIP., ECHEVERRIAP., KALAMBAHETIT. & CHONGSA-NGUAN M. 1988. Detection with monoclonal antibody of Salmonella typhi antigen 9 in specimens from patients. Journal of Clinical Microbiology 23: 1824-1830. CHAICUMPA W., RUANGKUNAWRN Y., KALAMBAHETIT., LIMAVONGPRANEE S., KITIK~~N V., KHUSMITHS., PUNGPAK S., CHONGSA-NGUANM. & SORNMANIS. 1991. Specific monoclonal antibodies to Opisthorchis viverrini. Iniernational Journalfor Parasitology 21: 969-974. EVANSH., BOURGEOIS C.S., COMERD. S. & KESCHAMARAS N. 1971. Biliary tract changes in opisthorchiasis. The American Journal of Tropical Medicine and Hygiene 21: 667-671. FAUST E. C., SAWITZ W., TOBIE J., OD~M V., PERES C & LINCOMED.R. 1939. Comparative efficiency of various techniques for the diagnosis of protozoa and helminth eggs in feces. Journal of Parasitology 25: 241-262. FELMEIERH., FELDHEIMW., RASP F. & BIENZLEU. 1981. Das krankheit-spektrnmvon flunchtlingen aus siidostasein. Deutsches Arzteblatt 17: 817-823. FLAVELL D.T., PATANAPANYA~ATK., LUCAS S.B. & VONGSANGNAK V. 1980. Opisthorchis viverrini: liver changes in golden hamsters maintained on high and low protein diets. Acta Tropica 37: 337-350. GALEN R. S. 1980. Predictive values and efficiency of laboratory testing. Pediatric Clinics of North America 27: 861-869. GREEN E. L. 1986. Immunological detection of parasite antigen in faeces. Parasitology Today 2: 198-200. HANVANICHM., VI~CIDI R., LAUGHONB. E., BARTLETTJ. G. & YOLKENR. H. 1985 Stool desorbing activity: a possible cause of false-positive reactions in competitive enzyme immunoassays. Journal of Clinical Microbiology 21: 184 188. HARINASUTA C. 1969. Opisthorchiasis in Thailand. In: Proceedings of the 4th Southeast Asian Seminar on Parasitology and Tropical Medicine, Schistosomiasis and Other Snail-transmitted Helminthiasis (Edited by HARINASUTA C.), pp. 253-275. Thai Watana Panich Press, Bangkok. HARINASUTA T., RIGANTI M. & BUNNAG D. 1984. Opisthorchis viverrini infection: pathogenesis and clinical features. Drug Research 34: 1167-I 169.

Research

Note

531

Drug Research 34: 12141217. HARLOW E. & LANE D. 1988. Antibodies: A Laboratory SIRISINHA S., CHAWENGKIRTIKUL R., SERMSWAN R., Munual. Cold Spring Harbor Laboratory, U.S.A. HOVI T., VAISANEN V., UKKONEN P. & B~NSWRFF C. H. VON AMORNPANT S., MONKOLSUK S. & PANYIM S. 1991. Detection of Opbthorchis viverrini by monoclonal anti1982. Solid phase enzyme immunoassay for rotavirus body-based ELISA and DNA hybridization. American antigen: faecal protease activity as a reason for false negative results. Journal of Virology Methods 5: 45-53. Journal of Tropical Medicine and Hygiene 44: 140-145. INDRAWATI I., CHAICIJM~A W., SETASU&WP. & RUANOK~JNAPORN STOLL N. R. 1923. An effective method of counting hookY. 1991. Studies on immunodiagnosis of human paragonworn eggs in feces. The American Journal of Hygiene 3: 59. VISCIDI R., LAUGHON B., HANVANICHM., BARTLE~~J. G. & imiasis and specific antigen of Paragonimus heterotremus. YOLKEN R. H. 1984. Improved enzyme immunoassays for International Journal for Parasitology 21: 395401. KINSLEY C. V., PRASHAD A. J. & PICKERING L.K. 1986. the detection of antigens in fecal specimens. Investigation and correction of interfering factors. Journal of ImmunoDetection of rotavirus in stool specimens with monological Methorls 67: 129-143. clonal and polyclonal antibody-based assay systems. Journal of Clinical Microbiology 23~ 897-900. YOLKEN R. H. 1985. Solid-phase enzyme immunoassays for the detection of microbial antigens in body fluids. In: LOWRY 0. H., ROSEBROUGHN. J., FARR A. L. & RANDALL R. Manual of Clinical Microbiology (Edited by LENNEI-TE E. J. 1951’. Protein measurement with Folin phenol reagent. H., BALOWSA., HAUSLERW. J. & SHAD~MYH.J.), pp. Journal of Biological Chemistry 24: 193-200. RALWMYOSP., BUNNAG D. & HARINASUTAT. 1984. Worms 947-957. American Society for Microbiology, recovered in stools following praziquantel treatment. Washington DC.